[PATCH] btrfs: zoned: don't submit orphaned extent buffers

On a zoned filesystem btree_writepages() can encounter a dirty metadata
extent buffer whose block group no longer exists. Submitting a write for
such a buffer maps it to a stale/removed block-group and leaves the folio
under writeback forever, hanging later in filemap_fdatawait_range(), for
example the iput(btree_inode) in close_ctree(), which then hangs unmount.

This is caused by btrfs_clear_buffer_dirty() not clearing the dirty bit of
a freed tree block but it sets EXTENT_BUFFER_ZONED_ZEROOUT and keeps the
buffer dirty so that it is still written out to keep the zone's
meta_write_pointer advancing sequentially. So a freed metadata block
legitimately stays dirty until that zero-write completes.

Dropping these buffers is safe: the block group is empty, so they are
stale, unreferenced, already-freed blocks. Once the zone is reset their
zero-write is unneeded. Instead of submitting a such a write, finish the
writeback immediately.

Reported-by: Shin'ichiro Kawasaki <[email protected]>
Fixes: 7db94301a980 ("btrfs: zoned: introduce block group context to btrfs_eb_write_context")
Cc: [email protected]
Signed-off-by: Johannes Thumshirn <[email protected]>
---
 fs/btrfs/extent_io.c | 16 ++++++++++++++--
 1 file changed, 14 insertions(+), 2 deletions(-)

diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 0edd532174fa..4a029ae719e9 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -2280,7 +2280,8 @@ static void prepare_eb_write(struct extent_buffer *eb)
 }
 
 static noinline_for_stack void write_one_eb(struct extent_buffer *eb,
-					    struct writeback_control *wbc)
+					    struct writeback_control *wbc,
+					    bool submit)
 {
 	struct btrfs_fs_info *fs_info = eb->fs_info;
 	struct btrfs_bio *bbio;
@@ -2310,6 +2311,12 @@ static noinline_for_stack void write_one_eb(struct extent_buffer *eb,
 		wbc_account_cgroup_owner(wbc, folio, range_len);
 		folio_unlock(folio);
 	}
+
+	if (!submit) {
+		btrfs_bio_end_io(bbio, BLK_STS_OK);
+		return;
+	}
+
 	/*
 	 * If the fs is already in error status, do not submit any writeback
 	 * but immediately finish it.
@@ -2397,6 +2404,8 @@ int btree_writepages(struct address_space *mapping, struct writeback_control *wb
 		struct extent_buffer *eb;
 
 		while ((eb = eb_batch_next(&batch)) != NULL) {
+			bool submit = true;
+
 			ctx.eb = eb;
 
 			ret = btrfs_check_meta_write_pointer(eb->fs_info, &ctx);
@@ -2411,6 +2420,9 @@ int btree_writepages(struct address_space *mapping, struct writeback_control *wb
 				continue;
 			}
 
+			if (btrfs_is_zoned(fs_info) && !ctx.zoned_bg)
+				submit = false;
+
 			if (!lock_extent_buffer_for_io(eb, wbc))
 				continue;
 
@@ -2420,7 +2432,7 @@ int btree_writepages(struct address_space *mapping, struct writeback_control *wb
 				btrfs_schedule_zone_finish_bg(ctx.zoned_bg, eb);
 				ctx.zoned_bg->meta_write_pointer += eb->len;
 			}
-			write_one_eb(eb, wbc);
+			write_one_eb(eb, wbc, submit);
 		}
 		nr_to_write_done = (wbc->nr_to_write <= 0);
 		eb_batch_release(&batch);
-- 
2.54.0